TikTok is striving to create a new version of its recommendation algorithm, specifically for its 170 million users in the United States. This project aims to make the U.S. platform independent from its Chinese parent company, ByteDance, potentially easing the concerns of American lawmakers who are considering banning the app. This effort, initiated by ByteDance before recent legislative actions started to intensify in Congress, led to the passage of a law in April that requires TikTok to divest its U.S. operations.

Sources, who chose to remain anonymous due to a lack of authorization, revealed that this separation of the source code could prepare the foundation for the eventual divestiture of TikTok’s assets in the U.S. However, the company has indicated that there are no immediate plans for such a move, reasserting previous statements that selling off its U.S. assets is not feasible. Initially, TikTok declined to comment on these issues. After the story was published, TikTok criticized the report as misleading and factually incorrect, though it did not clarify these claims. The company also referenced a portion of its federal lawsuit, arguing that the required divestiture is impractical within the prescribed timeframe and technologically unfeasible. In recent months, teams of engineers across the U.S. and China at ByteDance and TikTok have been laboring to separate millions of lines of code, focusing particularly on the recommendation algorithm. This strenuous task highlights the complexities of detaching TikTok’s U.S. operations from ByteDance. Despite facing numerous legal and technical challenges, TikTok continues its determined efforts to ensure its U.S. operations are autonomous, even though this might lead to some disparities in performance after the separation.